• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 63 subscribers
  • Views 898 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TPS65217: PMIC shuts off at 3.1V input when powered through VIN_BAT even if UVLO is set to 2.73V

EB
Prodigy 30 points
Part Number: TPS65217

Hello,

I've been trying to test the operating range of TPS65217C on battery input. Based on the datasheet, I believe I could go as low as 2.73V (UVLO[1:0]=0x00). However, PMIC tends to shutoff when the battery input voltage falls down to 3.1V. This behavior seems to be consistent on my custom board, Beaglebone Black and PocketBeagle. Is there any other parameter apart from UVLO that would force PMIC to shutoff prematurely?

Thanks

EB

  • 0
    TI__Mastermind 30310 points
    EB,

    Every time the board shuts down, the UVLO value will reset to the default (the main I2C register map is volatile). Are you sure it is being set every time you power-on the board?

    When I tested all of the UVLO values, I made changes to the EEPROM memory (non-volatile) so that the change was permanent. I was able to accurately measure the PMIC shutting off when the input supply decreased to less than 2.73V

    It would not be an easy test to run quickly unless you included the write to change UVLO[1:0] = 00b in the initialization routine of the processor.
  • 0 in reply to Brian Berner
    Prodigy 30 points
    Hello Brian,

    Thanks for your response.

    Yes, I understand that the PMIC registers reset to their default on POR.

    The sequence of steps I follow on every power up are given below:

    1. Set input BAT voltage to about 4V and power up the board.
    2. Once Linux boots, I write 0x65 to 0x0B register through I2C to disable password protection for DEFUVLO register (0x18).
    3. I then set DEFUVLO register (0x18) to 0x00 using i2cset
    4. I verify if DEFUVLO was actually altered by reading it through i2cget
    5. I gradually reduce input BAT voltage in steps of 0.1V to find out upto what voltage PMIC operates.

    After executing the above steps, I do observe that the PMIC shuts off at 3.1V instead of its default (i.e., 3.3V). However, I don't understand how or why it chooses 3.1V instead of 2.73V corresponding to DEFUVLO = 0x00? Am I missing something?
  • 0 in reply to EB
    TI__Mastermind 30310 points

    EB,

    As I said before, I have an EVM with DEFUVLO re-programmed with UVLO[1:0] = 00b

    Below are the results of a test where I set the start value on a source-meter to 3.8V and ramp down to 0V in 20mV steps. After triggering the source-meter, I pressed the push-button to power-on the PMIC. I am measuring LDO1, which is the first regulator to turn on and the last to turn off in the sequence.

    Ch2 = VBAT, Ch3 = LDO1 

    3.1V is marked with cursor (a), which happens at time t0-72ms

    2.73V is marked with cursor (b), which happens at time t0+868ms (almost a full second later). The PMIC shuts off shortly after the threshold is crossed (internal delays would account for this delay. However, this test was done with PWR_EN held low, which means on LDO1 turns on and all other regulators are waiting for PWR_EN to be set high before turning on. This is why it is easy for me to measure UVLO accurately.

    The reason you are seeing the PMIC turn off before 2.73V is quite simple: LDO2 = 3.3V, and the drop-out voltage of an LDO is typically 200mV with a significant load current. You are not seeing the PMIC enter UVLO, you are seeing the PMIC experience a fault condition.

    I was able to see LDO2 turn off in between 3.1V and 2.73V with no load, and adding a load would presumably make LDO2 fault sooner (3.1V being a reasonable value).

    Ch2 = VBAT, Ch3 = LDO2 

    The PGOOD signal will go low at the same time LDO2 shuts off and the power-down sequence will run as shown in "Figure 7. Default Power-Up Sequence" in the TPS65217 datasheet.

  • 0 in reply to Brian Berner
    Prodigy 30 points
    Thanks Brian. I was able to make PocketBeagle operate below 3V on BAT input by setting both LDO2 and LDO4 to 3V.
  • 0 in reply to EB
    TI__Mastermind 30310 points
    Perfect! Good to hear we successfully resolved your issue. Hope you have fun with the PocketBeagle now that this issue is resolved.